It’s funny, I had a similar “discussion” (more of an argument, actually…!) with a fellow on another board, recently, about passes and their usefulness. It really comes down to what works for you and your shop. In the vast majority of cases passes are an absolute necessity and a no-brainer. But there are times when they can actually decrease efficiency rather than increase it. Case in point, we just finished a 60+ shot fly-through for a client, 15 mins of footage in total. We’re a small shop (2 of us max at any one time, often just me) and I can tell you that there’s no way in hell we were going to render 4,5 or 6 or more passes x 60+ to achieve each shot. The reality was that our client was being sent stills along each path and signed those stills off “as-is”. So when they did sign one off, that was that and we sent the full shot to the farm. We knew that it would not only be possible but actually desirable to get these shots looking spot-on out of the box in the beauty pass and not have to construct 60+ comp scripts and render 6 or so passes. We spent many weeks (actually months, over time) refining and perfecting everything in 3d for the base renders. The disk space savings alone were worthwhile. In our case, it was very much a better option not to use passes. Of course we did end up using many subset render passes (fixed objects, new objects etc.) but that’s par for the course. And the usual post work (CC etc.).
Passes are all about efficiency and flexibility. When you don’t need the flexibility (because you’re doing it all in the base render and it makes sense to do things this way) all that’s left is efficiency. I think a lot of really small shops and one-man-bands use passes simply because they “ought to” without really sitting down and working out if they will actually be of any real benefit. In our case they would have slowed our production down to a grinding halt and cost an enormous amount in storage costs. Instead, we reinvested that time into R&D and reinvested the cost into the renderfarm, allowing us to hit a higher quality and send more shots out than originally planned for, so it worked very well. But we were in the very fortunate position of having enough time to really nail the shots in the beauty pass down to every last detail. It really just comes down to what works for you, on a particular project and probably even on a particular shot. Shot 001 might warrant passes, 002 might not. But when you’re in a large shop with a very streamlined and strict pipeline, you’re probably going to find that it’s either “use em every time” or “don’t”.
There’s also the crucial element of “creativity”. We weren’t afforded much of that for this job because every last detail was specified by the architects, so there was relatively little messing around finding nice tones, lighting arrangements etc. It was a case of “Here are the drawings, the lighting design, the furniture spec and layout - off you go”. When you’ve got the opportunity to play around with things like this passes can be fantastic. In the same way that you might use something like Magic Bullet to find a few options for general looks for a mood board.
Anyway that was a little off-topic in a way, but just thought I’d put it out there.
And as Jaco said (hi mate, by the way!), pass/layer… all the same thing really. Just terminology. I think most people say “passes” these days and know what you mean. When you say “layers” that does start to imply certain things. Passes is a nice, flexible, open term.
